US20160224253A1 - Memory System and Method for Delta Writes - Google Patents
Memory System and Method for Delta Writes Download PDFInfo
- Publication number
- US20160224253A1 US20160224253A1 US14/664,127 US201514664127A US2016224253A1 US 20160224253 A1 US20160224253 A1 US 20160224253A1 US 201514664127 A US201514664127 A US 201514664127A US 2016224253 A1 US2016224253 A1 US 2016224253A1
- Authority
- US
- United States
- Prior art keywords
- memory
- data
- stored
- memory system
- delta
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0608—Saving storage space on storage systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0638—Organizing or formatting or addressing of data
- G06F3/064—Management of blocks
- G06F3/0641—De-duplication techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0638—Organizing or formatting or addressing of data
- G06F3/0643—Management of files
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0673—Single storage device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0673—Single storage device
- G06F3/0679—Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0683—Plurality of storage devices
- G06F3/0688—Non-volatile semiconductor memory arrays
Definitions
- Deduplication refers to the process of identifying N duplicate copies of data in memory, physically storing only one copy of the data, and using a table to map the N logical block addresses for the data to the one physical address that actually stores the data.
- deduplication algorithms can be very complex and require a large amount of firmware and hardware overhead.
- a memory system receives a request to store data in the memory and determines whether the data requested to be stored in the memory is a modified version of data already stored in the memory. If it is, the memory system compares the data requested to be stored in the memory with the data already stored in the memory to identify differences between the data to be stored and the data already stored. The memory system then stores the identified differences in the memory, along with a table that maps the stored identified differences to corresponding locations in the data already stored in the memory.
- a memory system that detects a delta write triggering event, identifies delta data between data requested to be stored in the memory and previously-stored data, stores the delta data in the memory, and stores an entry in a delta table in the memory that integrates the delta data with the previously-stored data.
- a memory system is provided with a memory, a modification detection module configured to determine whether data to be stored in the memory is a modified version of data already stored in the memory, a differencing engine configured to compare the data to be stored in the memory with the data already stored in the memory to identify differences between the data to be stored and the data already stored, and a modification storing module configured to store the differences in the memory and to store a table in the memory that maps the differences to the data already stored in the memory.
- FIG. 1A is a block diagram of a non-volatile memory system of an embodiment.
- FIG. 1B is a block diagram illustrating an exemplary storage module of an embodiment.
- FIG. 1C is a block diagram illustrating a hierarchical storage system of an embodiment.
- FIG. 2A is a block diagram illustrating exemplary components of the controller of the non-volatile memory system illustrated in FIG. 1A according to an embodiment.
- FIG. 2B is a block diagram illustrating exemplary components of the non-volatile memory of the non-volatile memory storage system illustrated in FIG. 1A according to an embodiment.
- FIGS. 3A-3E are illustrations of data modifications of an embodiment.
- FIG. 4 is a diagram of an embodiment for using a hint from a host to determine whether data is a modified version of already-stored data.
- FIG. 5 is a diagram of an embodiment illustrating how differences in to-be-stored and already-stored data are identified.
- FIG. 6 is a diagram of an embodiment for using a fingerprinting technique to determine whether data is a modified version of already-stored data.
- FIG. 7 is a block diagram of a differencing engine of an embodiment.
- FIGS. 8, 9, and 10 are diagrams illustrating operations of a differencing engine of an embodiment.
- FIGS. 11A and 11B are diagrams illustrating a delta write operation of an embodiment.
- FIG. 12 is a flow chart illustrating a delta write operation of an embodiment.
- FIGS. 13A-13C are diagrams illustrating a compaction operation of an embodiment.
- FIG. 1A is a block diagram illustrating a non-volatile memory system according to an embodiment of the subject matter described herein.
- non-volatile memory system 100 includes a controller 102 and non-volatile memory that may be made up of one or more non-volatile memory die 104 .
- the term die refers to the collection of non-volatile memory cells, and associated circuitry for managing the physical operation of those non-volatile memory cells, that are formed on a single semiconductor substrate.
- Controller 102 interfaces with a host system and transmits command sequences for read, program, and erase operations to non-volatile memory die 104 .
- the controller 102 (which may be a flash memory controller) can take the form of processing circuitry, a microprocessor or processor, and a computer-readable medium that stores computer-readable program code (e.g., firmware) executable by the (micro)processor, logic gates, switches, an application specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller, for example.
- the controller 102 can be configured with hardware and/or firmware to perform the various functions described below and shown in the flow diagrams. Also, some of the components shown as being internal to the controller can also be stored external to the controller, and other components can be used. Additionally, the phrase “operatively in communication with” could mean directly in communication with or indirectly (wired or wireless) in communication with through one or more components, which may or may not be shown or described herein.
- a flash memory controller is a device that manages data stored on flash memory and communicates with a host, such as a computer or electronic device.
- a flash memory controller can have various functionality in addition to the specific functionality described herein.
- the flash memory controller can format the flash memory to ensure the memory is operating properly, map out bad flash memory cells, and allocate spare cells to be substituted for future failed cells. Some part of the spare cells can be used to hold firmware to operate the flash memory controller and implement other features.
- the flash memory controller can convert the logical address received from the host to a physical address in the flash memory.
- the flash memory controller can also perform various memory management functions, such as, but not limited to, wear leveling (distributing writes to avoid wearing out specific blocks of memory that would otherwise be repeatedly written to) and garbage collection (after a block is full, moving only the valid pages of data to a new block, so the full block can be erased and reused).
- wear leveling distributing writes to avoid wearing out specific blocks of memory that would otherwise be repeatedly written to
- garbage collection after a block is full, moving only the valid pages of data to a new block, so the full block can be erased and reused).
- Non-volatile memory die 104 may include any suitable non-volatile storage medium, including NAND flash memory cells and/or NOR flash memory cells.
- the memory cells can take the form of solid-state (e.g., flash) memory cells and can be one-time programmable, few-time programmable, or many-time programmable.
- the memory cells can also be single-level cells (SLC), multiple-level cells (MLC), triple-level cells (TLC), or use other memory cell level technologies, now known or later developed.
- the memory cells can be fabricated in a two-dimensional or three-dimensional fashion.
- the interface between controller 102 and non-volatile memory die 104 may be any suitable flash interface, such as Toggle Mode 200, 400, or 800.
- memory system 100 may be a card based system, such as a secure digital (SD) or a micro secure digital (micro-SD) card. In an alternate embodiment, memory system 100 may be part of an embedded memory system.
- SD secure digital
- micro-SD micro secure digital
- non-volatile memory system 100 (sometimes referred to herein as a storage module) includes a single channel between controller 102 and non-volatile memory die 104
- the subject matter described herein is not limited to having a single memory channel.
- 2, 4, 8 or more NAND channels may exist between the controller and the NAND memory device, depending on controller capabilities.
- more than a single channel may exist between the controller and the memory die, even if a single channel is shown in the drawings.
- FIG. 1B illustrates a storage module 200 that includes plural non-volatile memory systems 100 .
- storage module 200 may include a storage controller 202 that interfaces with a host and with storage system 204 , which includes a plurality of non-volatile memory systems 100 .
- the interface between storage controller 202 and non-volatile memory systems 100 may be a bus interface, such as a serial advanced technology attachment (SATA) or peripheral component interface express (PCIe) interface.
- Storage module 200 in one embodiment, may be a solid state drive (SSD), such as found in portable computing devices, such as laptop computers, and tablet computers.
- SSD solid state drive
- FIG. 1C is a block diagram illustrating a hierarchical storage system.
- a hierarchical storage system 250 includes a plurality of storage controllers 202 , each of which controls a respective storage system 204 .
- Host systems 252 may, access memories within the storage system via a bus interface.
- the bus interface may be an NVMe or fiber channel over Ethernet (FCoE) interface.
- FCoE fiber channel over Ethernet
- the system illustrated in FIG. 1C may be a rack mountable mass storage system that is accessible by multiple host computers, such as would be found in a data center or other location where mass storage is needed.
- FIG. 2A is a block diagram illustrating exemplary components of controller 102 in more detail.
- Controller 102 includes a front end module 108 that interfaces with a host, a back end module 110 that interfaces with the one or more non-volatile memory die 104 , and various other modules that perform functions which will now be described in detail.
- a module may take the form of a packaged functional hardware unit designed for use with other components, a portion of a program code (e.g., software or firmware) executable by a (micro)processor or processing circuitry that usually performs a particular function of related functions, or a self-contained hardware or software component that interfaces with a larger system, for example.
- a program code e.g., software or firmware
- Modules of the controller 102 may include a modification detection module 111 configured to determine whether data to be stored in the memory 104 is a modified version of data already stored in the memory 104 , a differencing engine 112 configured to compare the data to be stored in the memory 104 with the data already stored in the memory 104 to identify differences between the data to be stored and the data already stored, and a modification storing module 113 configured to store the differences in the memory 104 and to store a table in the memory 104 that maps the differences to the data already stored in the memory 104 .
- a modification detection module 111 configured to determine whether data to be stored in the memory 104 is a modified version of data already stored in the memory 104
- a differencing engine 112 configured to compare the data to be stored in the memory 104 with the data already stored in the memory 104 to identify differences between the data to be stored and the data already stored
- a modification storing module 113 configured to store the differences in the memory 104 and to store a table in the memory 104 that maps the
- a buffer manager/bus controller 114 manages buffers in random access memory (RAM) 116 and controls the internal bus arbitration of controller 102 .
- a read only memory (ROM) 118 stores system boot code. Although illustrated in FIG. 2A as located separately from the controller 102 , in other embodiments one or both of the RAM 116 and ROM 118 may be located within the controller. In yet other embodiments, portions of RAM and ROM may be located both within the controller 102 and outside the controller.
- Front end module 108 includes a host interface 120 and a physical layer interface (PITY) 122 that provide the electrical interface with the host or next level storage controller.
- PITY physical layer interface
- the choice of the type of host interface 120 can depend on the type of memory being used. Examples of host interfaces 120 include, but are not limited to, SATA, SATA Express, SAS, Fibre Channel, USB, PCIe, and NVMe.
- the host interface 120 typically facilitates transfer for data, control signals, and timing signals.
- Back end module 110 includes an error correction controller (ECC) engine 124 that encodes the data bytes received from the host, and decodes and error corrects the data bytes read from the non-volatile memory.
- ECC error correction controller
- a command sequencer 126 generates command sequences, such as program and erase command sequences, to be transmitted to non-volatile memory die 104 .
- a RAID (Redundant Array of Independent Drives) module 128 manages generation of RAID parity and recovery of failed data. The RAID parity may be used as an additional level of integrity protection for the data being written into the memory device 104 . In some cases, the RAID module 128 may be a part of the ECC engine 124 .
- a memory interface 130 provides the command sequences to non-volatile memory die 104 and receives status information from non-volatile memory die 104 .
- memory interface 130 may be a double data rate (DDR) interface, such as a Toggle Mode 200, 400, or 800 interface.
- DDR double data rate
- a flash control layer 132 controls the overall operation of back end module 110 .
- System 100 includes media management layer 138 , which performs wear leveling of memory cells of non-volatile memory die 104 .
- System 100 also includes other discrete components 140 , such as external electrical interfaces, external RAM, resistors, capacitors, or other components that may interface with controller 102 .
- one or more of the physical layer interface 122 , RAID module 128 , media management layer 138 and buffer management/bus controller 114 are optional components that are not necessary in the controller 102 .
- FIG. 2B is a block diagram illustrating exemplary components of non-volatile memory die 104 in more detail.
- Non-volatile memory die 104 includes peripheral circuitry 141 and non-volatile memory array 142 .
- Non-volatile memory array 142 includes the non-volatile memory cells used to store data.
- the non-volatile memory cells may be any suitable non-volatile memory cells, including NAND flash memory cells and/or NOR flash memory cells in a two dimensional and/or three dimensional configuration.
- Peripheral circuitry 141 includes a state machine 152 that provides status information to controller 102 .
- Non-volatile memory die 104 further includes a data cache 156 that caches data.
- delta writes provide more efficient ways of storing a modified version of data already stored in memory, which will be referred to herein as “delta writes.”
- the memory system 100 can use external or internal hints to derive that the operation is a modification and then determine the extent of the modification and whether to apply the write as a new write or as a delta write. More specifically, when the memory system 100 receives a request to store data in the memory 104 , the memory system 100 (e.g., the controller 102 or another component in the memory system 100 ) determines whether the data requested to be stored in the memory 104 is a modified version of data already stored in the memory 104 .
- the memory system 100 compares the data requested to be stored in the memory 104 with the data already stored in the memory 104 to identify differences between the data to be stored and the data already stored. The memory system 100 then stores the identified differences in the memory 104 , along with a table that maps the stored identified differences to corresponding locations in the data already stored in the memory 104 .
- any suitable granularity of “data” can be used.
- the “data requested to be stored” and the “data to be stored” can be an entire file, an “extent” (a sequential run of logical space), a string, or one or more minimum read/write units (e.g., a “codeword” with data and ECC bits).
- a “codeword” with data and ECC bits can be used.
- a “modified” version of data already stored in the memory can refer to any type of modification, including, but not limited to, a change, an addition, and/or a deletion.
- a modification may or may not change the overall length of the originally-stored data.
- FIGS. 3A-3E provide illustrations of various examples of modifications.
- the previously-stored data in the memory 104 (“old data”) is a data string
- the data to be stored in the memory 104 (“new data”) is a changed version of the data string.
- a section (Mod(X)) in the data string is being replaced with a section of the same size.
- the data to be stored in the memory 104 is of the same size as the previously-stored data, so the modification does not change the length of the data string.
- the modification is a deletion of the section. So, here, the modification results in a removal of bytes, reducing the overall length of the data string.
- the modification is an addition of a data. So, here, the modification results in the insertion of bytes, increasing the overall length of the data string.
- FIG. 3D shows a modification that is a combination of insertion and deletion. First, a section of data is being deleted, then another section of data is being changed.
- the modification is interpreted as a “deletion+insertion” instead of a “deletion+change,” with the deletion size being ⁇ Mod(X)+Mod(X), and the insertion size being Mod(X).
- the overall length of the data string is reduced.
- FIG. 3E there is a change to previously-stored data and an addition of data.
- the modification is interpreted as a “deletion+insertion” instead of a “deletion+change,” with the deletion size being Mod(X), and the insertion size being Mod(X)+Mod(X).
- the overall length of the data string is increased.
- the request to store a modified version of previously-stored data can come from a module internal to the storage system 100 or from a component external to the storage system 100 , such as a host.
- hosts include, but are not limited to, a mobile phone, a tablet computer, a digital media player, a game device, a personal digital assistant (PDA), a mobile (e.g., notebook, laptop) personal computer (PC), or a book reader.
- PDA personal digital assistant
- PC personal computer
- the first step is to determine if the delta write technique should even be applied at all (i.e., whether a delta write triggering event has occurred).
- the memory system 100 e.g., using the controller 102 , the modification detection module 111 , or another component in the memory system 100 ) determines whether data requested to be stored in the memory 104 is a modified version of data already stored in the memory 104 . This determination can be performed in any suitable manner.
- the memory system 100 can determine that data requested to be stored in the memory 104 is a modified version of data already stored in the memory 104 via an indication (or “hint”) received from the host sending the request to store the data.
- an indication or “hint”
- the host 400 includes a “mod write” flag in a command it sends to the memory system 100 to write data.
- this flag can be an indication that the write operation is a modification of an existing extent.
- the host 400 indicates via a hint through its operating system and input-output stack that this is a modification operation.
- the memory system 100 can then compare and derive the difference and, as discussed below, can figure out how best to store the data (e.g., either by storing the entire data as a complete set or as a modification).
- the memory system 100 can use the data stored in the memory 104 as an internal hint to derive whether or not the data requested to be stored in the memory 104 is a modified version of data already stored in the memory 104 .
- the memory system 100 can compare (e.g., using the differencing engine 112 ) part of the data to be stored (e.g., the first 4 KB (or other amount)) with corresponding parts of data already stored in the memory 104 (e.g., the first 4 KB of every stored file (or extent)) to see if there is a match. Looking at only part of the data avoids reading the entire data string from the memory 104 unless there is a match.
- the corresponding parts (“fingerprints”) of the stored data can be read from the memory 104 on-the-fly when the write request is received or can be read and pre-stored (e.g., in a hash table cached in RAM 116 or in the memory 104 ) before the request is received. If there is a match, it is likely that the data requested to be stored is a modified version of the matched stored data, and the delta write process can be performed. Otherwise, the data requested to be stored can be considered new data, and the delta write process does not need to be performed.
- the memory system 100 compares (e.g., using the controller 102 , the differencing engine 112 , or another component in the memory system 100 ) the data requested to be stored in the memory 104 with the data already stored in the memory 104 to identify differences between the data to be stored and the data already stored. This is shown generally in the block diagram of FIG. 5 .
- the controller 102 receives the data (Data String X′) to be written from the host 410 via the host interface 120 and stores it in a buffer 410 .
- the controller 102 also reads the data (Data String X) already stored in the memory 104 , passing it through the memory (flash) interface 130 , the ECC engine 124 , and the encryption (AES) engine 420 , and then stores it in a buffer 410 .
- the read from the memory 104 can occur in parallel to the transfer of data from the host 400 . If the operations are done in sequence, the read from the memory 104 can occur before or after the transfer of data from the host 400 .
- the controller 102 compares the two data strings to identify the differences (mod(X)) and stores only the identified differences in memory 104 , instead of the entire data string (Data String X′) received from the host 410 . (As shown in FIG. 6 , this comparing process can also be used to determine if incoming data matches “fingerprints” of the heads of strings of stored data (the process discussed above to determine if a delta write triggering event has occurred).)
- FIG. 7 shows the comparing operation in more detail and illustrates the use of the differencing engine 112 .
- the differencing engine 112 comprises a register interface 710 , a first direct memory access (DMA) controller 720 , an auto-buffer management (ABM) interface 730 , and a second DMA controller 740 .
- the first DMA controller 720 communicates (via an advanced high-performance (AHB) bus) with RAM 116 , which stores data streamed from the host 400 (via the host interface module (HIM) 120 and host-side DMA 750 ).
- HAB advanced high-performance
- the second DMA controller 740 communicates (via an AHB bus) with RAM 116 ′ (RAM 116 ′ can be part of or separate from RAM 116 ), which stores previously-stored data streamed from the memory (here, Flash) 104 via the flash interface module (FIM) 130 , ECC engine 114 , AES engine 420 , and flash-side DMA controller 760 .
- the differencing engine 112 compares the two data strings as they are streamed into the differencing engine 112 , and the identified differences are stored in the register interface 710 and, eventually provided to the microprocessor 780 (via the advanced peripheral bus (APB) bus), so the microprocessor 780 can store the identified difference in the memory 104 (a process which will be discussed in more detail below).
- the differencing engine 112 is implemented as hardware executing firmware or as an all-hardware implementation, as a general-purpose processor may be too slow to perform the comparison operation without adversely affecting the performance of the memory system 100 .
- the ABM module 770 can be used to throttle the streams of the incoming and/or stored data to avoid overflowing the RAM 116 , 116 ′.
- the memory system 100 stores (e.g., using the controller 103 , the modification storing module 113 , or some other component in the memory system 100 ) the identified differences in the memory 104 , along with a table (sometimes referred to herein as a “delta table”) that maps the stored identified differences to corresponding locations in the data already stored in the memory 104 .
- FIGS. 8-10 illustrate the operation how the mapping information can be generated.
- the general operation of the differencing engine 112 consists of parsing the two streams until a change (Mod(X)) is found.
- the differencing engine 112 can search both the new and old data streams with a limited extent search range to limit the size of the possible delta. In this way, the differences are stored only if they are less than the maximum extend size (i.e., only if Mod(x) ⁇ Max extent size), which can be configurable via a register by firmware. So, if the maximum extent size is 16 KB, any delta greater than that can be considered and written as a new cluster.
- the firmware can begin to back up and store the Mod(X) at the last 4 KB boundary.
- factors other than delta size can be used in determining whether to store only the identified differences or to store all the data requested to be stored. Such other factors include, but are not limited to, a number of times identified differences have been stored for the data already stored in the memory, the age of the data already stored in the memory, and whether or not a rollback feature (discussed below) is implemented.
- FIGS. 11A and 11B illustrate the use of the delta table discussed above.
- FIG. 11A shows the storing of the original data in the memory 104 .
- logical address A is assigned, and the data is written in the memory 104 .
- the host is attempting a second write to logical address A with a modification of the stored data string.
- the data string is broken into three parts in this example.
- the first part (A) is the unaffected part of the write.
- the second part (B) is the place of the written data that was modified.
- the last part (C) is the byproduct of changing the size of the second part (B) in the modified string.
- the flash translation layer (FTL) entry in the FTL table can still point to that part of the previously-stored data in the memory 104 .
- part B was the part of the data string that was changed, that part accounts for the FTL alignment boundaries. So, an entry is added in the delta table with information that maps the identified difference (part B) to the corresponding location in the previously-stored data, and the FTL entry in the FTL table for part B would point to the delta entry in the delta table instead of previously-stored part B.
- the FTL entry in the FTL table for part C can still point to the location of previously-stored part C.
- the FTL entry to the part C may need to go through an exception table if it has to be byte addressed instead of flash-management-unit (FMU) addressed.
- FMU flash-management-unit
- the delta table can be used to integrate/merge the delta data with the previously-stored data during a read operation, so that a request to read data string X will result in reading out original parts A and C, as well as the delta data (Mod(X)) instead of outdated part B.
- the use of the delta data results in reading out the modified data string instead of the original data string, even though only the deltas of the modified data string were actually written.
- mapping table is different from the delta table discussed herein. For example, in deduplication, an entry is written in the mapping table for every duplicate entry, so that the logical address for those duplication entries maps to the one physical location storing the data. In contrast, with a delta table, only information about the differences—not the matching data—are stored. This reduces the amount of writes needs to be made to the table, since writes are only made when there are differences.
- a time stamp can be stored along with the identified differences (deltas) to enable a tracking/rollback feature.
- the rollback feature can capture timestamps and retain old copies of the data. In this way, delta updates can become logs and be “unraveled” to make old data persistent for a period of time.
- a delta write is performed. These factors include, but are not limited to, a size of the delta, a number of times deltas have been stored for the data already stored in the memory, age of the delta data already stored in the memory, and whether a rollback feature is implemented.
- This alternative will be illustrated in conjunction with the flow chart 1200 in FIG. 12 .
- the memory system 100 determines if the rollback features is enabled (act 1220 ). If the rollback features is enabled, the write is recorded as a delta (act 1230 ).
- the memory system 100 determines if a bypass features is enabled to not reduce performance (act 1240 ). If it is, the write is recorded as a new write (act 1250 ); otherwise, the memory system 100 determines if a prior version of the data reached a threshold of data retention (DR) or read disturbs (RD), which would indicate that the old data is nearing scrub time (act 1260 ).
- DR data retention
- RD read disturbs
- the write is recorded as a new write, thereby re-writing the entire data set (act 1250 ); otherwise, the memory system 100 determines if a maximum number of deltas has been reached (act 1270 ). If it has, the write is recorded as a new write (act 1250 ); otherwise, the write is recorded as a delta (act 1230 ).
- the memory system 100 can collect a set of deltas and stored them together (e.g., in a page). (A delta can be as small as a single byte, and deltas can be collected as fragments, just like small write fragments.) Avoiding small update writes provides performance and endurance advantages for the memory system 100 and avoids inefficiencies at the host 400 . Collecting the deltas allows writes to be performed faster when a delta write hit occurs, but there can be a read performance penalty in having to coalesce a series of deltas. This alternative is shown in more detail in FIGS. 13A-13C .
- the delta fragments are not aligned to any logical boundary (be it a fragment of an ECC page (for example, 512 bytes) or an ECC page).
- the alignment is in bytes, and the length is in bytes.
- the delta data can be stored aligned to a logical boundary within the metapage.
- FIG. 13C (Derivation B.2), another savings is recognized by making the delta length a multiple of a constant size (instead of in bytes).
- any suitable type of memory can be used.
- Semiconductor memory devices include volatile memory devices, such as dynamic random access memory (“DRAM”) or static random access memory (“SRAM”) devices, non-volatile memory devices, such as resistive random access memory (“ReRAM”), electrically erasable programmable read only memory (“EEPROM”), flash memory (which can also be considered a subset of EEPROM), ferroelectric random access memory (“FRAM”), and magnetoresistive random access memory (“MRAM”), and other semiconductor elements capable of storing information.
- volatile memory devices such as dynamic random access memory (“DRAM”) or static random access memory (“SRAM”) devices
- non-volatile memory devices such as resistive random access memory (“ReRAM”), electrically erasable programmable read only memory (“EEPROM”), flash memory (which can also be considered a subset of EEPROM), ferroelectric random access memory (“FRAM”), and magnetoresistive random access memory (“MRAM”), and other semiconductor elements capable of storing information.
- ReRAM resistive random access memory
- the memory devices can be formed from passive and/or active elements, in any combinations.
- passive semiconductor memory elements include ReRAM device elements, which in some embodiments include a resistivity switching storage element, such as an anti-fuse, phase change material, etc., and optionally a steering element, such as a diode, etc.
- active semiconductor memory elements include EEPROM and flash memory device elements, which in some embodiments include elements containing a charge storage region, such as a floating gate, conductive nanoparticles, or a charge storage dielectric material.
- Multiple memory elements may be configured so that they are connected in series or so that each element is individually accessible.
- flash memory devices in a NAND configuration typically contain memory elements connected in series.
- a NAND memory array may be configured so that the array is composed of multiple strings of memory in which a string is composed of multiple memory elements sharing a single bit line and accessed as a group.
- memory elements may be configured so that each element is individually accessible, e.g., a NOR memory array.
- NAND and NOR memory configurations are exemplary, and memory elements may be otherwise configured.
- the semiconductor memory elements located within and/or over a substrate may be arranged in two or three dimensions, such as a two dimensional memory structure or a three dimensional memory structure.
- the semiconductor memory elements are arranged in a single plane or a single memory device level.
- memory elements are arranged in a plane (e.g., in an x-z direction plane) which extends substantially parallel to a major surface of a substrate that supports the memory elements.
- the substrate may be a wafer over or in which the layer of the memory elements are formed or it may be a carrier substrate which is attached to the memory elements after they are formed.
- the substrate may include a semiconductor such as silicon.
- the memory elements may be arranged in the single memory device level in an ordered array, such as in a plurality of rows and/or columns. However, the memory elements may be arrayed in non-regular or non-orthogonal configurations.
- the memory elements may each have two or more electrodes or contact lines, such as bit lines and word lines.
- a three dimensional memory array is arranged so that memory elements occupy multiple planes or multiple memory device levels, thereby forming a structure in three dimensions (i.e., in the x, y and z directions, where the y direction is substantially perpendicular and the x and z directions are substantially parallel to the major surface of the substrate).
- a three dimensional memory structure may be vertically arranged as a stack of multiple two dimensional memory device levels.
- a three dimensional memory array may be arranged as multiple vertical columns (e.g., columns extending substantially perpendicular to the major surface of the substrate, i.e., in the y direction) with each column having multiple memory elements in each column.
- the columns may be arranged in a two dimensional configuration, e.g., in an x-z plane, resulting in a three dimensional arrangement of memory elements with elements on multiple vertically stacked memory planes.
- Other configurations of memory elements in three dimensions can also constitute a three dimensional memory array.
- the memory elements may be coupled together to form a NAND string within a single horizontal (e.g., x-z) memory device levels.
- the memory elements may be coupled together to form a vertical NAND string that traverses across multiple horizontal memory device levels.
- Other three dimensional configurations can be envisioned wherein some NAND strings contain memory elements in a single memory level while other strings contain memory elements which span through multiple memory levels.
- Three dimensional memory arrays may also be designed in a NOR configuration and in a ReRAM configuration.
- a monolithic three dimensional memory array typically, one or more memory device levels are formed above a single substrate.
- the monolithic three dimensional memory array may also have one or more memory layers at least partially within the single substrate.
- the substrate may include a semiconductor such as silicon.
- the layers constituting each memory device level of the array are typically formed on the layers of the underlying memory device levels of the array.
- layers of adjacent memory device levels of a monolithic three dimensional memory array may be shared or have intervening layers between memory device levels.
- non-monolithic stacked memories can be constructed by forming memory levels on separate substrates and then stacking the memory levels atop each other. The substrates may be thinned or removed from the memory device levels before stacking, but as the memory device levels are initially formed over separate substrates, the resulting memory arrays are not monolithic three dimensional memory arrays. Further, multiple two dimensional memory arrays or three dimensional memory arrays (monolithic or non-monolithic) may be formed on separate chips and then packaged together to form a stacked-chip memory device.
- Associated circuitry is typically required for operation of the memory elements and for communication with the memory elements.
- memory devices may have circuitry used for controlling and driving memory elements to accomplish functions such as programming and reading.
- This associated circuitry may be on the same substrate as the memory elements and/or on a separate substrate.
- a controller for memory read-write operations may be located on a separate controller chip and/or on the same substrate as the memory elements.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
- This application claims priority to India Patent Application No. 313/MUM/2015, filed on Jan. 30, 2015, entitled “Memory System and Method for Delta Writes,” the entire disclosure of which is hereby incorporated by reference.
- Many memory systems treat a host request to write a file as a new write, meaning that the entire file is written in memory even if the file is a modified version of an already-stored file. Accordingly, the memory may contain multiple copies of the same file, each with slight modifications, which is a sub-optimal use of the memory. One way to address this issue is by using a technique known as “deduplication.” “Deduplication” refers to the process of identifying N duplicate copies of data in memory, physically storing only one copy of the data, and using a table to map the N logical block addresses for the data to the one physical address that actually stores the data. However, deduplication algorithms can be very complex and require a large amount of firmware and hardware overhead.
- Embodiments of the present invention are defined by the claims, and nothing in this section should be taken as a limitation on those claims.
- By way of introduction, the below embodiments relate to a memory system and method for delta writes. In one embodiment, a memory system receives a request to store data in the memory and determines whether the data requested to be stored in the memory is a modified version of data already stored in the memory. If it is, the memory system compares the data requested to be stored in the memory with the data already stored in the memory to identify differences between the data to be stored and the data already stored. The memory system then stores the identified differences in the memory, along with a table that maps the stored identified differences to corresponding locations in the data already stored in the memory.
- In another embodiment, a memory system is provided that detects a delta write triggering event, identifies delta data between data requested to be stored in the memory and previously-stored data, stores the delta data in the memory, and stores an entry in a delta table in the memory that integrates the delta data with the previously-stored data.
- In yet another embodiment, a memory system is provided with a memory, a modification detection module configured to determine whether data to be stored in the memory is a modified version of data already stored in the memory, a differencing engine configured to compare the data to be stored in the memory with the data already stored in the memory to identify differences between the data to be stored and the data already stored, and a modification storing module configured to store the differences in the memory and to store a table in the memory that maps the differences to the data already stored in the memory.
- Other embodiments are possible, and each of the embodiments can be used alone or together in combination. Accordingly, various embodiments will now be described with reference to the attached drawings.
-
FIG. 1A is a block diagram of a non-volatile memory system of an embodiment. -
FIG. 1B is a block diagram illustrating an exemplary storage module of an embodiment. -
FIG. 1C is a block diagram illustrating a hierarchical storage system of an embodiment. -
FIG. 2A is a block diagram illustrating exemplary components of the controller of the non-volatile memory system illustrated inFIG. 1A according to an embodiment. -
FIG. 2B is a block diagram illustrating exemplary components of the non-volatile memory of the non-volatile memory storage system illustrated inFIG. 1A according to an embodiment. -
FIGS. 3A-3E are illustrations of data modifications of an embodiment. -
FIG. 4 is a diagram of an embodiment for using a hint from a host to determine whether data is a modified version of already-stored data. -
FIG. 5 is a diagram of an embodiment illustrating how differences in to-be-stored and already-stored data are identified. -
FIG. 6 is a diagram of an embodiment for using a fingerprinting technique to determine whether data is a modified version of already-stored data. -
FIG. 7 is a block diagram of a differencing engine of an embodiment. -
FIGS. 8, 9, and 10 are diagrams illustrating operations of a differencing engine of an embodiment. -
FIGS. 11A and 11B are diagrams illustrating a delta write operation of an embodiment. -
FIG. 12 is a flow chart illustrating a delta write operation of an embodiment. -
FIGS. 13A-13C are diagrams illustrating a compaction operation of an embodiment. - As mentioned in the background section above, many memory systems treat a host request to write a file as a new write, meaning that the entire file is written in memory even if the file is a modified version of an already-stored file. To avoid wasting memory space with duplicate copies of the same data, a deduplication process can be used to identify N duplicate copies of data in memory and physically store only one copy of the data. However, deduplication algorithms can be very complex and require a large amount of firmware and hardware overhead. Further, even when duplicate data is identified, writes to the memory are still required, as the mapping table needs to be updated to show that the N logical block addresses for the data map to the one physical address that actually stores the data.
- The following embodiments provide more efficient ways of storing a modified version of data already stored in memory. Before turning to these and other embodiments, the following paragraphs provide a discussion of exemplary memory systems that can be used with these embodiments. Of course, these are just examples, and other suitable types of storage modules can be used.
- Memory systems suitable for use in implementing aspects of these embodiments are shown in
FIGS. 1A-1C .FIG. 1A is a block diagram illustrating a non-volatile memory system according to an embodiment of the subject matter described herein. Referring toFIG. 1A ,non-volatile memory system 100 includes acontroller 102 and non-volatile memory that may be made up of one or more non-volatile memory die 104. As used herein, the term die refers to the collection of non-volatile memory cells, and associated circuitry for managing the physical operation of those non-volatile memory cells, that are formed on a single semiconductor substrate.Controller 102 interfaces with a host system and transmits command sequences for read, program, and erase operations to non-volatile memory die 104. - The controller 102 (which may be a flash memory controller) can take the form of processing circuitry, a microprocessor or processor, and a computer-readable medium that stores computer-readable program code (e.g., firmware) executable by the (micro)processor, logic gates, switches, an application specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller, for example. The
controller 102 can be configured with hardware and/or firmware to perform the various functions described below and shown in the flow diagrams. Also, some of the components shown as being internal to the controller can also be stored external to the controller, and other components can be used. Additionally, the phrase “operatively in communication with” could mean directly in communication with or indirectly (wired or wireless) in communication with through one or more components, which may or may not be shown or described herein. - As used herein, a flash memory controller is a device that manages data stored on flash memory and communicates with a host, such as a computer or electronic device. A flash memory controller can have various functionality in addition to the specific functionality described herein. For example, the flash memory controller can format the flash memory to ensure the memory is operating properly, map out bad flash memory cells, and allocate spare cells to be substituted for future failed cells. Some part of the spare cells can be used to hold firmware to operate the flash memory controller and implement other features. In operation, when a host needs to read data from or write data to the flash memory, it will communicate with the flash memory controller. If the host provides a logical address to which data is to be read/written, the flash memory controller can convert the logical address received from the host to a physical address in the flash memory. (Alternatively, the host can provide the physical address.) The flash memory controller can also perform various memory management functions, such as, but not limited to, wear leveling (distributing writes to avoid wearing out specific blocks of memory that would otherwise be repeatedly written to) and garbage collection (after a block is full, moving only the valid pages of data to a new block, so the full block can be erased and reused).
- Non-volatile memory die 104 may include any suitable non-volatile storage medium, including NAND flash memory cells and/or NOR flash memory cells. The memory cells can take the form of solid-state (e.g., flash) memory cells and can be one-time programmable, few-time programmable, or many-time programmable. The memory cells can also be single-level cells (SLC), multiple-level cells (MLC), triple-level cells (TLC), or use other memory cell level technologies, now known or later developed. Also, the memory cells can be fabricated in a two-dimensional or three-dimensional fashion.
- The interface between
controller 102 and non-volatile memory die 104 may be any suitable flash interface, such asToggle Mode memory system 100 may be a card based system, such as a secure digital (SD) or a micro secure digital (micro-SD) card. In an alternate embodiment,memory system 100 may be part of an embedded memory system. - Although, in the example illustrated in
FIG. 1A , non-volatile memory system 100 (sometimes referred to herein as a storage module) includes a single channel betweencontroller 102 and non-volatile memory die 104, the subject matter described herein is not limited to having a single memory channel. For example, in some NAND memory system architectures (such as the ones shown inFIGS. 1B and 1C ), 2, 4, 8 or more NAND channels may exist between the controller and the NAND memory device, depending on controller capabilities. In any of the embodiments described herein, more than a single channel may exist between the controller and the memory die, even if a single channel is shown in the drawings. -
FIG. 1B illustrates astorage module 200 that includes pluralnon-volatile memory systems 100. As such,storage module 200 may include astorage controller 202 that interfaces with a host and withstorage system 204, which includes a plurality ofnon-volatile memory systems 100. The interface betweenstorage controller 202 andnon-volatile memory systems 100 may be a bus interface, such as a serial advanced technology attachment (SATA) or peripheral component interface express (PCIe) interface.Storage module 200, in one embodiment, may be a solid state drive (SSD), such as found in portable computing devices, such as laptop computers, and tablet computers. -
FIG. 1C is a block diagram illustrating a hierarchical storage system. Ahierarchical storage system 250 includes a plurality ofstorage controllers 202, each of which controls arespective storage system 204.Host systems 252 may, access memories within the storage system via a bus interface. In one embodiment, the bus interface may be an NVMe or fiber channel over Ethernet (FCoE) interface. In one embodiment, the system illustrated inFIG. 1C may be a rack mountable mass storage system that is accessible by multiple host computers, such as would be found in a data center or other location where mass storage is needed. -
FIG. 2A is a block diagram illustrating exemplary components ofcontroller 102 in more detail.Controller 102 includes afront end module 108 that interfaces with a host, aback end module 110 that interfaces with the one or more non-volatile memory die 104, and various other modules that perform functions which will now be described in detail. A module may take the form of a packaged functional hardware unit designed for use with other components, a portion of a program code (e.g., software or firmware) executable by a (micro)processor or processing circuitry that usually performs a particular function of related functions, or a self-contained hardware or software component that interfaces with a larger system, for example. Modules of thecontroller 102 may include amodification detection module 111 configured to determine whether data to be stored in thememory 104 is a modified version of data already stored in thememory 104, adifferencing engine 112 configured to compare the data to be stored in thememory 104 with the data already stored in thememory 104 to identify differences between the data to be stored and the data already stored, and amodification storing module 113 configured to store the differences in thememory 104 and to store a table in thememory 104 that maps the differences to the data already stored in thememory 104. These modules will be discussed in more detail below. - Referring again to modules of the
controller 102, a buffer manager/bus controller 114 manages buffers in random access memory (RAM) 116 and controls the internal bus arbitration ofcontroller 102. A read only memory (ROM) 118 stores system boot code. Although illustrated inFIG. 2A as located separately from thecontroller 102, in other embodiments one or both of theRAM 116 andROM 118 may be located within the controller. In yet other embodiments, portions of RAM and ROM may be located both within thecontroller 102 and outside the controller. -
Front end module 108 includes ahost interface 120 and a physical layer interface (PITY) 122 that provide the electrical interface with the host or next level storage controller. The choice of the type ofhost interface 120 can depend on the type of memory being used. Examples ofhost interfaces 120 include, but are not limited to, SATA, SATA Express, SAS, Fibre Channel, USB, PCIe, and NVMe. Thehost interface 120 typically facilitates transfer for data, control signals, and timing signals. -
Back end module 110 includes an error correction controller (ECC)engine 124 that encodes the data bytes received from the host, and decodes and error corrects the data bytes read from the non-volatile memory. Acommand sequencer 126 generates command sequences, such as program and erase command sequences, to be transmitted to non-volatile memory die 104. A RAID (Redundant Array of Independent Drives)module 128 manages generation of RAID parity and recovery of failed data. The RAID parity may be used as an additional level of integrity protection for the data being written into thememory device 104. In some cases, theRAID module 128 may be a part of theECC engine 124. Amemory interface 130 provides the command sequences to non-volatile memory die 104 and receives status information from non-volatile memory die 104. In one embodiment,memory interface 130 may be a double data rate (DDR) interface, such as aToggle Mode flash control layer 132 controls the overall operation ofback end module 110. - Additional components of
system 100 illustrated inFIG. 2A includemedia management layer 138, which performs wear leveling of memory cells of non-volatile memory die 104.System 100 also includes otherdiscrete components 140, such as external electrical interfaces, external RAM, resistors, capacitors, or other components that may interface withcontroller 102. In alternative embodiments, one or more of thephysical layer interface 122,RAID module 128,media management layer 138 and buffer management/bus controller 114 are optional components that are not necessary in thecontroller 102. -
FIG. 2B is a block diagram illustrating exemplary components of non-volatile memory die 104 in more detail. Non-volatile memory die 104 includesperipheral circuitry 141 andnon-volatile memory array 142.Non-volatile memory array 142 includes the non-volatile memory cells used to store data. The non-volatile memory cells may be any suitable non-volatile memory cells, including NAND flash memory cells and/or NOR flash memory cells in a two dimensional and/or three dimensional configuration.Peripheral circuitry 141 includes astate machine 152 that provides status information tocontroller 102. Non-volatile memory die 104 further includes adata cache 156 that caches data. - As mentioned above, many memory systems treat a host request to write a file as a new write, meaning that the entire file is written in memory even if the file is a modified version of an already-stored file. However, most of the disk space except for temporary space is fairly static. Small updates to existing non-compressed files are perceived by the host as new updates, and the host will normally write the entire file again. Complex deduplication algorithms can be used to identify N duplicate copies of data in memory and physically store only one copy of the data but writes to the memory are still required for the duplicate copies, as a mapping table needs to be updated to show that the N logical block addresses for the data map to the one physical address that actually stores the data.
- The following embodiments provide more efficient ways of storing a modified version of data already stored in memory, which will be referred to herein as “delta writes.” With a delta write, instead of writing the entire file again, the
memory system 100 can use external or internal hints to derive that the operation is a modification and then determine the extent of the modification and whether to apply the write as a new write or as a delta write. More specifically, when thememory system 100 receives a request to store data in thememory 104, the memory system 100 (e.g., thecontroller 102 or another component in the memory system 100) determines whether the data requested to be stored in thememory 104 is a modified version of data already stored in thememory 104. If it is, thememory system 100 compares the data requested to be stored in thememory 104 with the data already stored in thememory 104 to identify differences between the data to be stored and the data already stored. Thememory system 100 then stores the identified differences in thememory 104, along with a table that maps the stored identified differences to corresponding locations in the data already stored in thememory 104. Before turning to a detailed example of this embodiment, the following paragraphs will set forth various terms that will be used in the example. - First, any suitable granularity of “data” can be used. For example, the “data requested to be stored” and the “data to be stored” can be an entire file, an “extent” (a sequential run of logical space), a string, or one or more minimum read/write units (e.g., a “codeword” with data and ECC bits). Of course, these are merely examples, and other forms of data can be used.
- Further, a “modified” version of data already stored in the memory can refer to any type of modification, including, but not limited to, a change, an addition, and/or a deletion. A modification may or may not change the overall length of the originally-stored data.
FIGS. 3A-3E provide illustrations of various examples of modifications. InFIG. 3A , the previously-stored data in the memory 104 (“old data”) is a data string, and the data to be stored in the memory 104 (“new data”) is a changed version of the data string. Specifically, a section (Mod(X)) in the data string is being replaced with a section of the same size. Accordingly, the data to be stored in thememory 104 is of the same size as the previously-stored data, so the modification does not change the length of the data string. InFIG. 3B , the modification is a deletion of the section. So, here, the modification results in a removal of bytes, reducing the overall length of the data string. InFIG. 3C , the modification is an addition of a data. So, here, the modification results in the insertion of bytes, increasing the overall length of the data string.FIG. 3D shows a modification that is a combination of insertion and deletion. First, a section of data is being deleted, then another section of data is being changed. In this particular example, the modification is interpreted as a “deletion+insertion” instead of a “deletion+change,” with the deletion size being −Mod(X)+Mod(X), and the insertion size being Mod(X). Here, the overall length of the data string is reduced. InFIG. 3E , there is a change to previously-stored data and an addition of data. In this particular example, the modification is interpreted as a “deletion+insertion” instead of a “deletion+change,” with the deletion size being Mod(X), and the insertion size being Mod(X)+Mod(X). Here, the overall length of the data string is increased. - The request to store a modified version of previously-stored data can come from a module internal to the
storage system 100 or from a component external to thestorage system 100, such as a host. Examples of hosts include, but are not limited to, a mobile phone, a tablet computer, a digital media player, a game device, a personal digital assistant (PDA), a mobile (e.g., notebook, laptop) personal computer (PC), or a book reader. - With examples of “data” and “modification” having been discussed, the following paragraphs provide an exemplary implementation of a delta write technique of an embodiment. As explained above, the first step is to determine if the delta write technique should even be applied at all (i.e., whether a delta write triggering event has occurred). To do this, the memory system 100 (e.g., using the
controller 102, themodification detection module 111, or another component in the memory system 100) determines whether data requested to be stored in thememory 104 is a modified version of data already stored in thememory 104. This determination can be performed in any suitable manner. For example, thememory system 100 can determine that data requested to be stored in thememory 104 is a modified version of data already stored in thememory 104 via an indication (or “hint”) received from the host sending the request to store the data. This is shown inFIG. 4 , where thehost 400 includes a “mod write” flag in a command it sends to thememory system 100 to write data. (In the NVMe protocol, this flag can be an indication that the write operation is a modification of an existing extent.) Here, thehost 400 indicates via a hint through its operating system and input-output stack that this is a modification operation. (The use of hints reduces the overhead needed in thememory system 100 for the delta write operation.) Thememory system 100 can then compare and derive the difference and, as discussed below, can figure out how best to store the data (e.g., either by storing the entire data as a complete set or as a modification). - As another example, instead of or in addition to using an indication from the
host 400, thememory system 100 can use the data stored in thememory 104 as an internal hint to derive whether or not the data requested to be stored in thememory 104 is a modified version of data already stored in thememory 104. For example, thememory system 100 can compare (e.g., using the differencing engine 112) part of the data to be stored (e.g., the first 4 KB (or other amount)) with corresponding parts of data already stored in the memory 104 (e.g., the first 4 KB of every stored file (or extent)) to see if there is a match. Looking at only part of the data avoids reading the entire data string from thememory 104 unless there is a match. The corresponding parts (“fingerprints”) of the stored data can be read from thememory 104 on-the-fly when the write request is received or can be read and pre-stored (e.g., in a hash table cached inRAM 116 or in the memory 104) before the request is received. If there is a match, it is likely that the data requested to be stored is a modified version of the matched stored data, and the delta write process can be performed. Otherwise, the data requested to be stored can be considered new data, and the delta write process does not need to be performed. - Assuming the delta write process is to be performed, the
memory system 100 then compares (e.g., using thecontroller 102, thedifferencing engine 112, or another component in the memory system 100) the data requested to be stored in thememory 104 with the data already stored in thememory 104 to identify differences between the data to be stored and the data already stored. This is shown generally in the block diagram ofFIG. 5 . - As shown in
FIG. 5 , thecontroller 102 receives the data (Data String X′) to be written from thehost 410 via thehost interface 120 and stores it in abuffer 410. Thecontroller 102 also reads the data (Data String X) already stored in thememory 104, passing it through the memory (flash)interface 130, theECC engine 124, and the encryption (AES)engine 420, and then stores it in abuffer 410. The read from thememory 104 can occur in parallel to the transfer of data from thehost 400. If the operations are done in sequence, the read from thememory 104 can occur before or after the transfer of data from thehost 400. Thecontroller 102 then compares the two data strings to identify the differences (mod(X)) and stores only the identified differences inmemory 104, instead of the entire data string (Data String X′) received from thehost 410. (As shown inFIG. 6 , this comparing process can also be used to determine if incoming data matches “fingerprints” of the heads of strings of stored data (the process discussed above to determine if a delta write triggering event has occurred).) -
FIG. 7 shows the comparing operation in more detail and illustrates the use of thedifferencing engine 112. As shown inFIG. 7 , thedifferencing engine 112 comprises aregister interface 710, a first direct memory access (DMA)controller 720, an auto-buffer management (ABM)interface 730, and asecond DMA controller 740. Thefirst DMA controller 720 communicates (via an advanced high-performance (AHB) bus) withRAM 116, which stores data streamed from the host 400 (via the host interface module (HIM) 120 and host-side DMA 750). Thesecond DMA controller 740 communicates (via an AHB bus) withRAM 116′ (RAM 116′ can be part of or separate from RAM 116), which stores previously-stored data streamed from the memory (here, Flash) 104 via the flash interface module (FIM) 130,ECC engine 114,AES engine 420, and flash-side DMA controller 760. Thedifferencing engine 112 compares the two data strings as they are streamed into thedifferencing engine 112, and the identified differences are stored in theregister interface 710 and, eventually provided to the microprocessor 780 (via the advanced peripheral bus (APB) bus), so themicroprocessor 780 can store the identified difference in the memory 104 (a process which will be discussed in more detail below). - In one embodiment, the
differencing engine 112 is implemented as hardware executing firmware or as an all-hardware implementation, as a general-purpose processor may be too slow to perform the comparison operation without adversely affecting the performance of thememory system 100. Also, due to possible delay involved in comparing the data streams, theABM module 770 can be used to throttle the streams of the incoming and/or stored data to avoid overflowing theRAM - After the differences in the two data streams are identified, the
memory system 100 stores (e.g., using the controller 103, themodification storing module 113, or some other component in the memory system 100) the identified differences in thememory 104, along with a table (sometimes referred to herein as a “delta table”) that maps the stored identified differences to corresponding locations in the data already stored in thememory 104.FIGS. 8-10 illustrate the operation how the mapping information can be generated. As shown inFIG. 8 , the general operation of thedifferencing engine 112 consists of parsing the two streams until a change (Mod(X)) is found. However, it may also be desired to record the location in the original data that is affected by the modification. For example, as shown inFIG. 9 , thememory system 100 can store, in addition to the modified data itself, the starting offset of the modification, as well as the length of the insertion (if applicable), the length of the deletion (if applicable), the length of change (if applicable), and a flag to signal that a difference was detection (an interrupt request (IRQ) can also be used). So, in addition to the modified data itself, the memory system can store the corresponding location information in the following format (although other formats can be used): Offset=xyz; Insert=0; Delete=0; Mod=sizeof(Mod(X)). This information can then be stored as an entry in the delta table (e.g., inRAM 116 or the memory 104). (Use of the delta table will be discussed in more detail below). - Depending on the size of the difference between the two data streams, it may be preferred to store the entirety of the new data stream instead of just the differences. So, the
differencing engine 112 can search both the new and old data streams with a limited extent search range to limit the size of the possible delta. In this way, the differences are stored only if they are less than the maximum extend size (i.e., only if Mod(x)<Max extent size), which can be configurable via a register by firmware. So, if the maximum extent size is 16 KB, any delta greater than that can be considered and written as a new cluster.FIG. 10 illustrates an example in which the differences are greater than the maximum extend size (i.e., Mod(x)>=Max extent size). Here, thedifferencing engine 112 stops the comparison prematurely, and a hardware flag can be set that indicates that the extent limit was reached (e.g., Offset=xyz; Insert=N/A; Delete=N/A; Mod=N/A). The firmware can begin to back up and store the Mod(X) at the last 4 KB boundary. It should be noted that factors other than delta size can be used in determining whether to store only the identified differences or to store all the data requested to be stored. Such other factors include, but are not limited to, a number of times identified differences have been stored for the data already stored in the memory, the age of the data already stored in the memory, and whether or not a rollback feature (discussed below) is implemented. - Returning to the drawings,
FIGS. 11A and 11B illustrate the use of the delta table discussed above.FIG. 11A shows the storing of the original data in thememory 104. During this first write, logical address A is assigned, and the data is written in thememory 104. At a later time, the host is attempting a second write to logical address A with a modification of the stored data string. Under the delta write technique, only the identified difference in the data would be stored in thememory 104. As shown inFIG. 11B , the data string is broken into three parts in this example. The first part (A) is the unaffected part of the write. The second part (B) is the place of the written data that was modified. The last part (C) is the byproduct of changing the size of the second part (B) in the modified string. - As shown at the bottom of
FIG. 11B , since there was no change to part A of the originally-stored data, the flash translation layer (FTL) entry in the FTL table can still point to that part of the previously-stored data in thememory 104. However, since part B was the part of the data string that was changed, that part accounts for the FTL alignment boundaries. So, an entry is added in the delta table with information that maps the identified difference (part B) to the corresponding location in the previously-stored data, and the FTL entry in the FTL table for part B would point to the delta entry in the delta table instead of previously-stored part B. Assuming that the modification to part B did not change its length, the FTL entry in the FTL table for part C can still point to the location of previously-stored part C. However, if the modification to part B did change its length (e.g., due to an insertion or deletion), the FTL entry to the part C may need to go through an exception table if it has to be byte addressed instead of flash-management-unit (FMU) addressed. - The delta table can be used to integrate/merge the delta data with the previously-stored data during a read operation, so that a request to read data string X will result in reading out original parts A and C, as well as the delta data (Mod(X)) instead of outdated part B. Thus, the use of the delta data results in reading out the modified data string instead of the original data string, even though only the deltas of the modified data string were actually written.
- It should be noted that while prior deduplication processes use a mapping table, that mapping table is different from the delta table discussed herein. For example, in deduplication, an entry is written in the mapping table for every duplicate entry, so that the logical address for those duplication entries maps to the one physical location storing the data. In contrast, with a delta table, only information about the differences—not the matching data—are stored. This reduces the amount of writes needs to be made to the table, since writes are only made when there are differences.
- There are many alternatives that can be used with these embodiments. For example, a time stamp can be stored along with the identified differences (deltas) to enable a tracking/rollback feature. The rollback feature can capture timestamps and retain old copies of the data. In this way, delta updates can become logs and be “unraveled” to make old data persistent for a period of time.
- In another alternative, several factors can weigh into whether or not a delta write is performed. These factors include, but are not limited to, a size of the delta, a number of times deltas have been stored for the data already stored in the memory, age of the delta data already stored in the memory, and whether a rollback feature is implemented. This alternative will be illustrated in conjunction with the
flow chart 1200 inFIG. 12 . As shown inFIG. 12 , when delta detection is triggered (act 1210), thememory system 100 determines if the rollback features is enabled (act 1220). If the rollback features is enabled, the write is recorded as a delta (act 1230). If the rollback features is not enabled (e.g., via drive, drive partition, or by input-output request), thememory system 100 determines if a bypass features is enabled to not reduce performance (act 1240). If it is, the write is recorded as a new write (act 1250); otherwise, thememory system 100 determines if a prior version of the data reached a threshold of data retention (DR) or read disturbs (RD), which would indicate that the old data is nearing scrub time (act 1260). If it has (e.g., if the data has reached its “end of life”), the write is recorded as a new write, thereby re-writing the entire data set (act 1250); otherwise, thememory system 100 determines if a maximum number of deltas has been reached (act 1270). If it has, the write is recorded as a new write (act 1250); otherwise, the write is recorded as a delta (act 1230). - As another alternative, instead of storing deltas as they occur, the
memory system 100 can collect a set of deltas and stored them together (e.g., in a page). (A delta can be as small as a single byte, and deltas can be collected as fragments, just like small write fragments.) Avoiding small update writes provides performance and endurance advantages for thememory system 100 and avoids inefficiencies at thehost 400. Collecting the deltas allows writes to be performed faster when a delta write hit occurs, but there can be a read performance penalty in having to coalesce a series of deltas. This alternative is shown in more detail inFIGS. 13A-13C . - In
FIG. 13A (Derivation A), the delta fragments are not aligned to any logical boundary (be it a fragment of an ECC page (for example, 512 bytes) or an ECC page). The alignment is in bytes, and the length is in bytes. InFIG. 13B (Derivation B.1), for a slight savings of space in storing the physical address in the delta table, the delta data can be stored aligned to a logical boundary within the metapage. InFIG. 13C (Derivation B.2), another savings is recognized by making the delta length a multiple of a constant size (instead of in bytes). - Finally, as mentioned above, any suitable type of memory can be used. Semiconductor memory devices include volatile memory devices, such as dynamic random access memory (“DRAM”) or static random access memory (“SRAM”) devices, non-volatile memory devices, such as resistive random access memory (“ReRAM”), electrically erasable programmable read only memory (“EEPROM”), flash memory (which can also be considered a subset of EEPROM), ferroelectric random access memory (“FRAM”), and magnetoresistive random access memory (“MRAM”), and other semiconductor elements capable of storing information. Each type of memory device may have different configurations. For example, flash memory devices may be configured in a NAND or a NOR configuration.
- The memory devices can be formed from passive and/or active elements, in any combinations. By way of non-limiting example, passive semiconductor memory elements include ReRAM device elements, which in some embodiments include a resistivity switching storage element, such as an anti-fuse, phase change material, etc., and optionally a steering element, such as a diode, etc. Further by way of non-limiting example, active semiconductor memory elements include EEPROM and flash memory device elements, which in some embodiments include elements containing a charge storage region, such as a floating gate, conductive nanoparticles, or a charge storage dielectric material.
- Multiple memory elements may be configured so that they are connected in series or so that each element is individually accessible. By way of non-limiting example, flash memory devices in a NAND configuration (NAND memory) typically contain memory elements connected in series. A NAND memory array may be configured so that the array is composed of multiple strings of memory in which a string is composed of multiple memory elements sharing a single bit line and accessed as a group. Alternatively, memory elements may be configured so that each element is individually accessible, e.g., a NOR memory array. NAND and NOR memory configurations are exemplary, and memory elements may be otherwise configured.
- The semiconductor memory elements located within and/or over a substrate may be arranged in two or three dimensions, such as a two dimensional memory structure or a three dimensional memory structure.
- In a two dimensional memory structure, the semiconductor memory elements are arranged in a single plane or a single memory device level. Typically, in a two dimensional memory structure, memory elements are arranged in a plane (e.g., in an x-z direction plane) which extends substantially parallel to a major surface of a substrate that supports the memory elements. The substrate may be a wafer over or in which the layer of the memory elements are formed or it may be a carrier substrate which is attached to the memory elements after they are formed. As a non-limiting example, the substrate may include a semiconductor such as silicon.
- The memory elements may be arranged in the single memory device level in an ordered array, such as in a plurality of rows and/or columns. However, the memory elements may be arrayed in non-regular or non-orthogonal configurations. The memory elements may each have two or more electrodes or contact lines, such as bit lines and word lines.
- A three dimensional memory array is arranged so that memory elements occupy multiple planes or multiple memory device levels, thereby forming a structure in three dimensions (i.e., in the x, y and z directions, where the y direction is substantially perpendicular and the x and z directions are substantially parallel to the major surface of the substrate).
- As a non-limiting example, a three dimensional memory structure may be vertically arranged as a stack of multiple two dimensional memory device levels. As another non-limiting example, a three dimensional memory array may be arranged as multiple vertical columns (e.g., columns extending substantially perpendicular to the major surface of the substrate, i.e., in the y direction) with each column having multiple memory elements in each column. The columns may be arranged in a two dimensional configuration, e.g., in an x-z plane, resulting in a three dimensional arrangement of memory elements with elements on multiple vertically stacked memory planes. Other configurations of memory elements in three dimensions can also constitute a three dimensional memory array.
- By way of non-limiting example, in a three dimensional NAND memory array, the memory elements may be coupled together to form a NAND string within a single horizontal (e.g., x-z) memory device levels. Alternatively, the memory elements may be coupled together to form a vertical NAND string that traverses across multiple horizontal memory device levels. Other three dimensional configurations can be envisioned wherein some NAND strings contain memory elements in a single memory level while other strings contain memory elements which span through multiple memory levels. Three dimensional memory arrays may also be designed in a NOR configuration and in a ReRAM configuration.
- Typically, in a monolithic three dimensional memory array, one or more memory device levels are formed above a single substrate. Optionally, the monolithic three dimensional memory array may also have one or more memory layers at least partially within the single substrate. As a non-limiting example, the substrate may include a semiconductor such as silicon. In a monolithic three dimensional array, the layers constituting each memory device level of the array are typically formed on the layers of the underlying memory device levels of the array. However, layers of adjacent memory device levels of a monolithic three dimensional memory array may be shared or have intervening layers between memory device levels.
- Then again, two dimensional arrays may be formed separately and then packaged together to form a non-monolithic memory device having multiple layers of memory. For example, non-monolithic stacked memories can be constructed by forming memory levels on separate substrates and then stacking the memory levels atop each other. The substrates may be thinned or removed from the memory device levels before stacking, but as the memory device levels are initially formed over separate substrates, the resulting memory arrays are not monolithic three dimensional memory arrays. Further, multiple two dimensional memory arrays or three dimensional memory arrays (monolithic or non-monolithic) may be formed on separate chips and then packaged together to form a stacked-chip memory device.
- Associated circuitry is typically required for operation of the memory elements and for communication with the memory elements. As non-limiting examples, memory devices may have circuitry used for controlling and driving memory elements to accomplish functions such as programming and reading. This associated circuitry may be on the same substrate as the memory elements and/or on a separate substrate. For example, a controller for memory read-write operations may be located on a separate controller chip and/or on the same substrate as the memory elements.
- One of skill in the art will recognize that this invention is not limited to the two dimensional and three dimensional exemplary structures described but cover all relevant memory structures within the spirit and scope of the invention as described herein and as understood by one of skill in the art.
- It is intended that the foregoing detailed description be understood as an illustration of selected forms that the invention can take and not as a definition of the invention. It is only the following claims, including all equivalents, that are intended to define the scope of the claimed invention. Finally, it should be noted that any aspect of any of the preferred embodiments described herein can be used alone or in combination with one another.
Claims (38)
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
IN313/MUM/2015 | 2015-01-30 | ||
IN313MU2015 | 2015-01-30 |
Publications (2)
Publication Number | Publication Date |
---|---|
US20160224253A1 true US20160224253A1 (en) | 2016-08-04 |
US9904472B2 US9904472B2 (en) | 2018-02-27 |
Family
ID=56554250
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US14/664,127 Active 2035-06-06 US9904472B2 (en) | 2015-01-30 | 2015-03-20 | Memory system and method for delta writes |
Country Status (1)
Country | Link |
---|---|
US (1) | US9904472B2 (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10318202B2 (en) * | 2017-03-20 | 2019-06-11 | Via Technologies, Inc. | Non-volatile memory apparatus and data deduplication method thereof |
US10956346B1 (en) * | 2017-01-13 | 2021-03-23 | Lightbits Labs Ltd. | Storage system having an in-line hardware accelerator |
US20210255810A1 (en) * | 2017-09-08 | 2021-08-19 | Samsung Electronics Co., Ltd. | Storage device including nonvolatile memory device and controller, controller and operating method of nonvolatile memory device |
CN113687791A (en) * | 2021-09-07 | 2021-11-23 | 天津津航计算技术研究所 | Memory data fast reading device based on pre-reading and branch judgment mechanism |
CN113687792A (en) * | 2021-09-07 | 2021-11-23 | 天津津航计算技术研究所 | Method for quickly reading data of memory |
US20230063730A1 (en) * | 2021-08-31 | 2023-03-02 | Lemon Inc. | Storage engine for hybrid data processing |
US11841845B2 (en) | 2021-08-31 | 2023-12-12 | Lemon Inc. | Data consistency mechanism for hybrid data processing |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109240605B (en) * | 2018-08-17 | 2020-05-19 | 华中科技大学 | Rapid repeated data block identification method based on 3D stacked memory |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070130232A1 (en) * | 2005-11-22 | 2007-06-07 | Therrien David G | Method and apparatus for efficiently storing and managing historical versions and replicas of computer data files |
US20080010378A1 (en) * | 2003-07-10 | 2008-01-10 | Parker James A | Collaborative File Update System |
US8001215B2 (en) * | 2005-11-29 | 2011-08-16 | Microsoft Corporation | Unlimited history store for navigational web applications |
US20110246869A1 (en) * | 2010-04-06 | 2011-10-06 | Xerox Corporation | Method and system for processing documents through document history encapsulation |
US20130013557A1 (en) * | 2011-07-08 | 2013-01-10 | Peter Kunath | Method of updating a database of a navigation device and navigation device |
US20130055231A1 (en) * | 2011-08-31 | 2013-02-28 | Avaya Inc. | System and method for incremental software installation |
US20130191342A1 (en) * | 2007-03-05 | 2013-07-25 | Sachhin Sreedharan | Using versioning to back up multiple versions of a stored object |
-
2015
- 2015-03-20 US US14/664,127 patent/US9904472B2/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080010378A1 (en) * | 2003-07-10 | 2008-01-10 | Parker James A | Collaborative File Update System |
US20070130232A1 (en) * | 2005-11-22 | 2007-06-07 | Therrien David G | Method and apparatus for efficiently storing and managing historical versions and replicas of computer data files |
US8001215B2 (en) * | 2005-11-29 | 2011-08-16 | Microsoft Corporation | Unlimited history store for navigational web applications |
US20130191342A1 (en) * | 2007-03-05 | 2013-07-25 | Sachhin Sreedharan | Using versioning to back up multiple versions of a stored object |
US20150154234A1 (en) * | 2007-03-05 | 2015-06-04 | Emc Corporation | Using versioning to back up multiple versions of a stored object |
US20110246869A1 (en) * | 2010-04-06 | 2011-10-06 | Xerox Corporation | Method and system for processing documents through document history encapsulation |
US20130013557A1 (en) * | 2011-07-08 | 2013-01-10 | Peter Kunath | Method of updating a database of a navigation device and navigation device |
US20130055231A1 (en) * | 2011-08-31 | 2013-02-28 | Avaya Inc. | System and method for incremental software installation |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10956346B1 (en) * | 2017-01-13 | 2021-03-23 | Lightbits Labs Ltd. | Storage system having an in-line hardware accelerator |
US11256431B1 (en) | 2017-01-13 | 2022-02-22 | Lightbits Labs Ltd. | Storage system having a field programmable gate array |
US10318202B2 (en) * | 2017-03-20 | 2019-06-11 | Via Technologies, Inc. | Non-volatile memory apparatus and data deduplication method thereof |
US20210255810A1 (en) * | 2017-09-08 | 2021-08-19 | Samsung Electronics Co., Ltd. | Storage device including nonvolatile memory device and controller, controller and operating method of nonvolatile memory device |
US11693605B2 (en) * | 2017-09-08 | 2023-07-04 | Samsung Electronics Co., Ltd. | Storage device including nonvolatile memory device and controller, controller and operating method of nonvolatile memory device |
US20230063730A1 (en) * | 2021-08-31 | 2023-03-02 | Lemon Inc. | Storage engine for hybrid data processing |
US11789936B2 (en) * | 2021-08-31 | 2023-10-17 | Lemon Inc. | Storage engine for hybrid data processing |
US11841845B2 (en) | 2021-08-31 | 2023-12-12 | Lemon Inc. | Data consistency mechanism for hybrid data processing |
CN113687791A (en) * | 2021-09-07 | 2021-11-23 | 天津津航计算技术研究所 | Memory data fast reading device based on pre-reading and branch judgment mechanism |
CN113687792A (en) * | 2021-09-07 | 2021-11-23 | 天津津航计算技术研究所 | Method for quickly reading data of memory |
Also Published As
Publication number | Publication date |
---|---|
US9904472B2 (en) | 2018-02-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10282252B2 (en) | RAID storage device and method of management thereof | |
US9904472B2 (en) | Memory system and method for delta writes | |
US10157004B2 (en) | Storage system and method for recovering data corrupted in a host memory buffer | |
US10002042B2 (en) | Systems and methods of detecting errors during read operations and skipping word line portions | |
US10901620B2 (en) | Storage system and method for thin provisioning | |
KR101447786B1 (en) | Power interrupt management | |
JP5649742B2 (en) | Transaction log restore | |
US9990146B2 (en) | Apparatus and method of data sequencing | |
US9812209B2 (en) | System and method for memory integrated circuit chip write abort indication | |
US11262928B2 (en) | Storage system and method for enabling partial defragmentation prior to reading in burst mode | |
US11875036B2 (en) | Computing system including host and storage system and having increased write performance | |
US11550658B1 (en) | Storage system and method for storing logical-to-physical address table entries in a codeword in volatile memory | |
US11880603B2 (en) | Storage system and method for preserving overwritten data to allow a future rollback | |
US11847343B2 (en) | Storage system and method for non-blocking coherent re-writes | |
US20230384966A1 (en) | Storage System and Method for Data Placement in Zoned Storage | |
US11314428B1 (en) | Storage system and method for detecting and utilizing wasted space using a file system | |
US11520695B2 (en) | Storage system and method for automatic defragmentation of memory | |
US11573893B2 (en) | Storage system and method for validation of hints prior to garbage collection | |
US11301168B2 (en) | Storage system and method for user-defined data archiving | |
US20240176501A1 (en) | Data Storage Device and Method for Swap Defragmentation | |
US20240256180A1 (en) | Data Storage Device and Method for Host-Assisted Deferred Defragmentation and System Handling | |
US20240143509A1 (en) | Data Storage Device and Method for Handling Write Commands in Zoned Storage | |
US11650758B2 (en) | Data storage device and method for host-initiated cached read to recover corrupted data within timeout constraints | |
US11809331B1 (en) | Storage system and method for avoiding header to improve parity | |
US20210373789A1 (en) | Storage System, Host, and Method for Optimizing Storage of a Sequence of Images |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: SANDISK TECHNOLOGIES INC., TEXAS Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:TUERS, DANIEL E.;MANOHAR, ABHIJEET;HAHN, JUDAH GAMLIEL;REEL/FRAME:035220/0402 Effective date: 20150317 |
|
AS | Assignment |
Owner name: SANDISK TECHNOLOGIES LLC, TEXAS Free format text: CHANGE OF NAME;ASSIGNOR:SANDISK TECHNOLOGIES INC;REEL/FRAME:038812/0954 Effective date: 20160516 |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
MAFP | Maintenance fee payment |
Free format text: PAYMENT OF MAINTENANCE FEE, 4TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1551); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY Year of fee payment: 4 |